Singapore Launches World’s First AI Testing Framework and Toolkit

The Future of AI:
Responsible and Trustworthy
Publications
In the past few years, we have published papers about AI diagnosis on conferences and journals in software engineering and security, e.g., ICSE, S&P, CAV, TSE, and FSE.
These papers cover fields of trustworthy AI, including robustness, fairness, security and explainability. In addition, we won two ACM SIGSOFT Distinguished Paper Award (ICSE 2018, ICSE 2020) and one ACM SIGSOFT Research Highlights (2020).
QuoTe: Quality-oriented Testing for Deep Learning Systems, TOSEM, Feb 2023
Adversarial Attacks and Mitigation for Anomaly Detectors of Cyber-Physical Systems, IJCIP, Sep 2021
RobOT: Robustness-Oriented Testing for Deep Learning Systems, ICSE, May 2021
Adversarial Sample Detection for Deep Neural Network through Model Mutation Testing, ICSE, May 2019
Towards Optimal Concolic Testing, ICSE, May 2018
Robustness
Fairness
FairRec: Fairness Testing for Deep Recommender Systems, ISSTA, Jul 2023
View >>
TestSGD: Interpretable Testing of Neural Networks Against Subtle Group Discrimination, TOSEM, Apr 2023
Adaptive Fairness Improvement based Causality Analysis, FSE, Nov 2022
Automatic Fairness Testing of Neural Classifiers Through Adversarial Sampling, TSE, Aug 2021
White-box Fairness Testing through Adversarial Sampling, ICSE, June 2020
Security
Explainability
Semantic-based Neural Network Repair, ISSTA, Jul 2023
View >>
Which Neural Network Makes More Explainable Decisions? An Approach towards Measuring Explainability, ASE-J, Nov 2022
ExAIs: Executable AI Semantics, ICSE, Jul 2022
Towards Interpreting Recurrent Neural Network through Probabilistic Abstraction, ASE, Sep 2020
Copyright
Copy, Right? A Testing Framework for Copyright Protection of Deep Learning Models, S&P, May 2022